commit 346274bcba9d4c8499e9f54eeb5b6ff398c5c9cf Author: ModelHub XC Date: Sat May 9 12:25:51 2026 +0800 初始化项目，由ModelHub XC社区提供模型 Model: worthdoing/Phi-4-mini-GGUF Source: Original Platform diff --git a/.gitattributes b/.gitattributes new file mode 100644 index 0000000..59e40f9 --- /dev/null +++ b/.gitattributes @@ -0,0 +1,42 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +phi-4-mini-Q4_K_M-worthdoing.gguf filter=lfs diff=lfs merge=lfs -text +phi-4-mini-Q5_K_M-worthdoing.gguf filter=lfs diff=lfs merge=lfs -text +phi-4-mini-Q8_0-worthdoing.gguf filter=lfs diff=lfs merge=lfs -text +phi-4-mini-Q3_K_M-worthdoing.gguf filter=lfs diff=lfs merge=lfs -text +phi-4-mini-Q4_K_S-worthdoing.gguf filter=lfs diff=lfs merge=lfs -text +phi-4-mini-Q5_K_S-worthdoing.gguf filter=lfs diff=lfs merge=lfs -text +phi-4-mini-Q6_K-worthdoing.gguf filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000..45881f7 --- /dev/null +++ b/README.md @@ -0,0 +1,135 @@ +--- +language: +- en +- fr +- multilingual +license: apache-2.0 +tags: +- gguf +- quantized +- mac +- apple-silicon +- local-inference +- worthdoing +base_model: microsoft/Phi-4-mini-instruct +quantized_by: worthdoing +pipeline_tag: text-generation +--- + +

+ worthdoing +

+

Author: Simon-Pierre Boucher

+ +

+ GGUF + Parameters + Apple Silicon + License + worthdoing +

+

+ Q4_K_M + Q5_K_M + Q8_0 +

+ +# Phi-4-mini - GGUF Quantized by worthdoing + +> Quantized for local Mac inference (Apple Silicon / Metal) by **worthdoing** + +## About + +This is a GGUF quantized version of [Phi-4-mini](https://huggingface.co/microsoft/Phi-4-mini-instruct), optimized for running locally on Apple Silicon Macs with `llama.cpp`, `Ollama`, or `LM Studio`. + +- **Original model:** [microsoft/Phi-4-mini-instruct](https://huggingface.co/microsoft/Phi-4-mini-instruct) +- **Parameters:** 3.8B +- **Quantized by:** worthdoing +- **Pipeline:** corelm-model v1.0 + +## Description + +Microsoft's reasoning powerhouse in a tiny package. + +## Available Quantizations + +| File | Quant | BPW | Size | Use Case | +|------|-------|-----|------|----------| +| `phi-4-mini-Q4_K_M-worthdoing.gguf` | Q4_K_M | 4.58 | ~2.0 GB | **Recommended** - Best quality/size ratio | +| `phi-4-mini-Q5_K_M-worthdoing.gguf` | Q5_K_M | 5.33 | ~2.4 GB | Higher quality, still fast | +| `phi-4-mini-Q8_0-worthdoing.gguf` | Q8_0 | 7.96 | ~3.5 GB | Near-original quality | + +## How to Use + +### With Ollama +```bash +# Create a Modelfile +cat > Modelfile <<'MODELEOF' +FROM ./phi-4-mini-Q4_K_M-worthdoing.gguf +MODELEOF + +ollama create phi-4-mini -f Modelfile +ollama run phi-4-mini +``` + +### With llama.cpp +```bash +llama-cli -m phi-4-mini-Q4_K_M-worthdoing.gguf -p "Your prompt here" -ngl 99 +``` + +### With LM Studio +1. Download the GGUF file +2. Open LM Studio -> My Models -> Import +3. Select the GGUF file and start chatting + +## Quantization Method + +Our quantization pipeline (**corelm-model v1.0**) follows a rigorous multi-step process to ensure maximum quality and compatibility: + +### Step 1 — Download & Validation +- Model weights are downloaded from HuggingFace Hub in **SafeTensors** format (`.safetensors`) +- Legacy formats (`.bin`, `.pt`) are excluded to ensure clean, verified weights +- Tokenizer, configuration, and all metadata are preserved + +### Step 2 — Conversion to GGUF F16 Baseline +- The original model is converted to **GGUF format at FP16 precision** using `convert_hf_to_gguf.py` from [llama.cpp](https://github.com/ggml-org/llama.cpp) +- This lossless baseline preserves the full original model quality +- Architecture-specific tensors (attention, FFN, embeddings, MoE routing) are mapped to their GGUF equivalents + +### Step 3 — K-Quant Quantization +- The F16 baseline is quantized using `llama-quantize` with **k-quant methods** +- K-quants use a mixed-precision approach: more important layers (attention, output) retain higher precision, while less sensitive layers (FFN) are compressed more aggressively +- Each quantization level offers a different quality/size tradeoff: + +| Method | Bits per Weight | Strategy | +|--------|----------------|----------| +| **Q4_K_M** | ~4.58 bpw | Mixed 4/5-bit. Attention & output layers use Q5_K, FFN layers use Q4_K. Best balance of quality and size. | +| **Q5_K_M** | ~5.33 bpw | Mixed 5/6-bit. Attention & output layers use Q6_K, FFN layers use Q5_K. Higher quality with moderate size increase. | +| **Q8_0** | ~7.96 bpw | Uniform 8-bit. All layers quantized to 8-bit. Near-lossless quality, largest file size. | + +### Step 4 — Metadata Injection +- Custom metadata is embedded directly in each GGUF file: + - `general.quantized_by`: worthdoing + - `general.quantization_version`: corelm-1.0 +- This ensures full traceability and provenance of every quantized file + +### Tools & Environment +- **llama.cpp**: Used for both conversion and quantization — the industry-standard open-source LLM inference engine +- **Target platform**: Apple Silicon Macs (M1/M2/M3/M4) with Metal GPU acceleration +- **Inference runtimes**: Compatible with `llama.cpp`, `Ollama`, `LM Studio`, `koboldcpp`, and any GGUF-compatible runtime + +## Recommended Hardware + +| Quant | Min RAM | Recommended | +|-------|---------|-------------| +| Q4_K_M | 4 GB | Mac with 8 GB+ RAM | +| Q5_K_M | 4 GB | Mac with 8 GB+ RAM | +| Q8_0 | 4 GB | Mac with 8 GB+ RAM | + +## Tags + +`general`, `reasoning`, `coding`, `math` + +--- + +*Quantized with corelm-model pipeline by **worthdoing** on 2026-04-17* diff --git a/phi-4-mini-Q3_K_M-worthdoing.gguf b/phi-4-mini-Q3_K_M-worthdoing.gguf new file mode 100644 index 0000000..8ea2917 --- /dev/null +++ b/phi-4-mini-Q3_K_M-worthdoing.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:465401c2153e72a3cd4eaa47fb34bc2216c9e262dd3d45ac70b1473309b32918 +size 2121464704 diff --git a/phi-4-mini-Q4_K_M-worthdoing.gguf b/phi-4-mini-Q4_K_M-worthdoing.gguf new file mode 100644 index 0000000..dcd0c2d --- /dev/null +++ b/phi-4-mini-Q4_K_M-worthdoing.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5f84c3ca3834e73f742081dd90e1a96f41a7eabab6accec858c1ac9d7e385191 +size 2493840256 diff --git a/phi-4-mini-Q4_K_S-worthdoing.gguf b/phi-4-mini-Q4_K_S-worthdoing.gguf new file mode 100644 index 0000000..63086ce --- /dev/null +++ b/phi-4-mini-Q4_K_S-worthdoing.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dbb01329402107fd43bb4118a3e40581c574377251c91f6b9d5a890dd4bf06a2 +size 2345597824 diff --git a/phi-4-mini-Q5_K_M-worthdoing.gguf b/phi-4-mini-Q5_K_M-worthdoing.gguf new file mode 100644 index 0000000..9dbc5c7 --- /dev/null +++ b/phi-4-mini-Q5_K_M-worthdoing.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1ae27d1599d01524385acb7999f0c984754f6dd9a5cf05403c0ef7aaae4614e3 +size 2814704512 diff --git a/phi-4-mini-Q5_K_S-worthdoing.gguf b/phi-4-mini-Q5_K_S-worthdoing.gguf new file mode 100644 index 0000000..222e5c2 --- /dev/null +++ b/phi-4-mini-Q5_K_S-worthdoing.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc0bd058adad5db66219f1ae4afd0e320fffd066ae0258b73ad27814087283b3 +size 2727803776 diff --git a/phi-4-mini-Q6_K-worthdoing.gguf b/phi-4-mini-Q6_K-worthdoing.gguf new file mode 100644 index 0000000..d9f9c42 --- /dev/null +++ b/phi-4-mini-Q6_K-worthdoing.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a796fc5bfaa7d6d0e0a77f1998b8b48d8db40bb0edb7131d28dc1a22a0d7177a +size 3155622784 diff --git a/phi-4-mini-Q8_0-worthdoing.gguf b/phi-4-mini-Q8_0-worthdoing.gguf new file mode 100644 index 0000000..317c6d9 --- /dev/null +++ b/phi-4-mini-Q8_0-worthdoing.gguf @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4026867af70b1471a71aca853dac19442c5dd4e799aa8e9d089bca97cad109cf +size 4084610944